J. Parker Mitchell, Mark E. Dean, Grant Bruer, James S. Plank and Garett S. Rose

July, 2018

ICONS: International Conference on Neuromorphic Systems

https://ornlcda.github.io/icons2018/

View Article

Abstract

Following from the original Dynamic Adaptive Neural Network Array (DANNA) model, we propose a new digital neuromorphic architecture named DANNA 2. Through this paper, we introduce our new hardware design and software simulator, and we explain how DANNA2 can improve network density, training convergence, and element performance as compared to the DANNA model. We propose two network arrays types catering to FPGA and VLSI use cases. Using simulation results for both control and classiication problems, we show improved training convergence, network density, and simulation performance.

Citation Information

Text

author          J. P. Mitchell and M. E. Dean and G. Bruer and J. S. Plank and G. S. Rose
title           {DANNA 2}: Dynamic Adaptive Neural Network Arrays
booktitle       International Conference on Neuromorphic Computing Systems
publisher       ACM
address         Knoxville, TN
month           July
year            2018
doi             10.1145/3229884.3229894
where           https://dl.acm.org/citation.cfm?id=3229894

Bibtex

@INPROCEEDINGS{mdb:18:dda,
    author = "J. P. Mitchell and M. E. Dean and G. Bruer and J. S. Plank and G. S. Rose",
    title = "{DANNA 2}: Dynamic Adaptive Neural Network Arrays",
    booktitle = "International Conference on Neuromorphic Computing Systems",
    publisher = "ACM",
    address = "Knoxville, TN",
    month = "July",
    year = "2018",
    doi = "10.1145/3229884.3229894",
    where = "https://dl.acm.org/citation.cfm?id=3229894"
}